Project Summary/Abstract Hunger is typically elicited by negative energy balance, and causes a state of increased motivation to seek out, work for, and eat food. Hunger biases attention toward food-associated cues (e.g. candy bar wrappers) so that calorie-dense foods can be found and consumed in order to restore energy balance. Enhanced behavioral sensitivity to food cues remains a major obstacle to weight-loss programs involving food restriction, and can even persist in satiated individuals suffering from obesity or eating disorders. Despite the clinical importance of this phenomenon, the cellular and circuit mechanisms by which hunger biases cognitive processing towards food-predicting cues remain largely unknown. A key brain area known to integrate information about internal bodily states such as hunger with external sensory cues to drive goal-directed behavior is the insular cortex (IC). Neuroimaging studies in humans have consistently found that hunger-dependent increases in the incentive value of visual food cues correlates with increased food-cue-evoked responses in IC. In rodents, an intact IC is critical for learned food-predicting cues to induce food-seeking behavior, potentially due to its role in the retrieval of the incentive value of these cues. The overarching goal of this proposal is to define the neural pathways by which hunger selectively enhances responses to food cues in IC. One promising starting point for this pathway is the set of hypothalamic agouti-related protein (AgRP) neurons that integrates interoceptive signals of negative energy balance. Hunger-related behaviors are restored in sated mice by activation of these neurons. Moreover, hunger-related enhancement of food-cue responses in IC are restored in sated human subjects by systemic injection of ghrelin, a hunger-stimulating hormone that activates AgRP neurons. We will combine reversible manipulation of AgRP neuron activity with a new imaging approach we developed for long- term imaging of the activity of individual neurons in mouse IC across slowly-changing motivational states. In Aim 1, we will test whether visual food cue responses in specific subsets of IC neurons are (i) selectively enhanced by food restriction, (ii) strongly attenuated by satiety, and (iii) restored by chemogenetic activation of AgRP neurons. In Aim 2, we will determine whether basolateral amygdala axons in IC (BLA?IC) are a necessary source of hunger-dependent food-cue information, via long-term imaging and optogenetic silencing of BLA?IC axons. In Aim 3, we will test the hypothesis that AgRP neurons projecting to the paraventricular thalamus (AgRP?PVT) mediate hunger-dependent modulation of food-cue responses in IC by inhibiting PVT inputs to BLA, using innovative circuit mapping techniques in brain slices and in vivo, together with optogenetic stimulation and silencing of AgRP?PVT neurons. This work should greatly advance our understanding of the mechanisms by which hunger exerts its potent effects on processing of food-related cues in IC, ultimately driving cravings, excessive eating, and obesity. More generally, these experiments establish a framework for understanding how the needs of the body drive flexible, goal-oriented processing of sensory information in IC.